Transactions on Emerging Telecommunications Technologies最新文献

{"title":"A Novel Adaptive Extreme Learning Machine for Traffic Prediction and Multipath Routing Framework in Software Defined Networks With Hybrid Optimization Approach for Smart Hotel Applications","authors":"Illuru Rajasekhar,&nbsp;M. Monisha","doi":"10.1002/ett.70272","DOIUrl":"https://doi.org/10.1002/ett.70272","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Problem Statement</h3>\u0000 \u0000 <p>The revolutionary growth of software-defined networking (SDN) has provided a flexible framework to design and improve network management. In a wide range of networks, traffic congestion remains a major challenge. When handling massive amounts of data, it can easily lead to scalability issues due to the rapid network growth, which negatively impacts network performance. Therefore, traffic prediction becomes a quite challenging task. In addition, SDN has proven successful in various applications within wireless communication systems. For enabling better data transmission, efficient routing is essential. During the routing process, energy consumption and link breakage often increase, which limits overall network performance.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methodology</h3>\u0000 \u0000 <p>A new traffic prediction and multipath routing model in SDN is developed based on machine learning techniques. The machine learning approach is utilized to develop an effective traffic prediction and multipath routing framework in the SDN system, considering flow rule space and Quality-of-Service constraints. Initially, the traffic present in the network is predicted using an Adaptive Extreme Learning Machine (A-ELM), whose parameters are tuned using the proposed Hybrid Position of Sheep Flock and Tunicate Swarm (HP-SFTS) algorithm. Here, routing performance is improved through the HP-SFTS, which effectively minimizes both the volume of routed traffic and the cost of communication path routing. In performance validation, the developed model accurately traces network traffic and also demonstrates resilience to noise in the training data.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>From the comparative analysis, the developed HP-SFTS-A-ELM model achieved scores of 37.25, 10.979, and 1387.6 in terms of root mean square error, mean absolute error, and mean squared error, respectively.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Implications of the Study</h3>\u0000 \u0000 <p>Considering the use of SDN in traffic prediction and multipath routing, this approach is primarily applicable in areas such as data center management, traffic engineering, and network slicing. SDN helps to enhance network performance by transmitting data through less congested routes, and it offers a better computational efficiency rate compared to classical techniques in different experimental analyses.</p>\u0000 </section>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MobStream: A Reinforcement-Driven Mobile Streaming Methodology Over Multipath QUIC","authors":"Yan Cui,&nbsp;Zongzheng Liang,&nbsp;Zicong Huang,&nbsp;Peng Guo,&nbsp;Shijie Jia","doi":"10.1002/ett.70264","DOIUrl":"https://doi.org/10.1002/ett.70264","url":null,"abstract":"<p>With the rapid growth of mobile communications, using mobile devices such as smartphones has become a trend. However, the limited uplink of the cellular network constrains the uploading quality. By integrating the link capacity of both cellular networks and WiFi, concurrent multipath transmission, such as multipath QUIC (MP-QUIC), becomes a promising solution for alleviating the uplink bottleneck issue. This paper proposes MobStream, a novel reinforcement learning-driven solution based on MP-QUIC. MobStream maximizes streaming capacity by fully utilizing the bandwidth of both WiFi and cellular networks through MP-QUIC. To address the issue of reduced performance caused by the differences in path quality between WiFi and cellular networks, MobStream incorporates partial reliability and a layered coding scheme, paving the way to adapt to varied network conditions without harming the bandwidth utilization. We formulate the concurrent transmission problem as a stochastic optimization task and demonstrate its solution using reinforcement learning methods. Furthermore, to ensure fairness in transmission among other single-path protocols, we further introduced a fairness factor to the reinforcement learning method. Extensive experiments demonstrate that MobStream outperforms state-of-the-art solutions in terms of bitrate, packet loss, and delay.</p>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ett.70264","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"V2X Fusion Communication Framework Based on VANETS Collaborative Autonomous Driving","authors":"Jinhua Yu,&nbsp;Guang Mei","doi":"10.1002/ett.70263","DOIUrl":"https://doi.org/10.1002/ett.70263","url":null,"abstract":"<div>\u0000 \u0000 <p>The advancement of collaborative autonomous driving relies on robust and efficient data exchange between vehicles and surrounding infrastructure. Vehicle-to-everything (V2X) fusion communication frameworks, built upon vehicular ad hoc networks (VANETs), enable the integration of heterogeneous data sources to enhance environmental perception and decision-making. However, practical implementation faces significant challenges due to communication interruptions inherent in dynamic VANET environments, leading to incomplete cooperative perception and increased safety risks. To address these challenges, this research proposes a V2X fusion communication framework, incorporating communication-interruption-aware cooperative perception, to ensure reliable information exchange for autonomous vehicles operating in collaborative scenarios. The framework leverages historical cooperation information to compensate for missing data caused by communication disruptions. Furthermore, a communication stochastic temporal convolutional networks (STCN) prediction model is introduced to extract critical features under varying network conditions, enhancing predictive accuracy for lost information. The data were collected from an open-source platform, which includes multi-agent sensor data (LiDAR, radar, and camera), global positioning system (GPS), and timestamped V2X messages simulating realistic vehicular traffic and environmental conditions under varying communication qualities. Packet drop rates were emulated to reflect real-world VANET communication inconsistencies. Additionally, knowledge distillation techniques provide targeted supervision to the predictive model, while curriculum learning strategies stabilize the training process under complex VANET scenarios. The results of the experiments prove that the proposed framework enhanced the perception reliability, and collaborative performance, communication reliability, decreased latency, enhanced obstacle detection accuracy, and decreased error results, including MAE (0.11) and MSE (0.12). This VANET communication architecture is a fusion-based framework that provides reliable, efficient, and safe data-driven collaboration within a group of autonomous vehicles.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning-Enhanced DDoS Attack Detection and Mitigation in VANET Infrastructure","authors":"T. Gayathri,&nbsp;S. Uma Maheswari,&nbsp;S. Ponni Alias Sathya,&nbsp;T. Satyanarayana Murthy,&nbsp;Pramoda Patro","doi":"10.1002/ett.70262","DOIUrl":"https://doi.org/10.1002/ett.70262","url":null,"abstract":"<div>\u0000 \u0000 <p>Vehicular ad-hoc networks (VANETs) are crucial for road safety, traffic management, and intelligent transportation systems, but they are vulnerable to Distributed Denial of Service (DDoS) attacks, which can severely disrupt communication between vehicles and Roadside Units (RSUs). Traditional DDoS detection methods in VANETs are often inefficient due to reliance on centralized architectures and handcrafted features. To address these challenges, we propose the Hybrid Deep Learning with Federated Learning (HDL-FL) framework, which leverages Convolutional Neural Networks (CNNs) to capture spatial and temporal traffic patterns. By utilizing Federated Learning, HDL-FL enables distributed, privacy-preserving training across RSUs and vehicles while reducing communication overhead. Experimental evaluations in simulated VANET environments show that HDL-FL achieves a 94% improvement in accuracy, a 30% reduction in false positives, and a 99% increase in attack detection rate while also reducing communication overhead by 6.5 s and latency by 160 ms. The framework offers a scalable, robust, and privacy-preserving solution for securing next-generation Vehicle-to-Everything (V2X) infrastructures, outperforming traditional models in terms of spatio-temporal accuracy and scalability. For performance validation, the HDL-FL framework is compared with baseline models, including traditional machine learning approaches such as Support Vector Machine, AI, and IoT.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Efficient and Secure WBAN Based on Optimal Privacy Preservation Scheme With Deep Learning and Blockchain Technology","authors":"Balasubramanian Chandra,&nbsp;Subramanian Kanaga Suba Raja,&nbsp;Suresh Sudha","doi":"10.1002/ett.70266","DOIUrl":"https://doi.org/10.1002/ett.70266","url":null,"abstract":"<div>\u0000 \u0000 <p>Securing the trustworthiness, privacy, and legitimacy of shared medical data in Wireless Body Area Network (WBAN) is a primary concern. Hence, a blockchain technology-based secure medical data storage scheme is developed in this paper. This developed model includes four primary phases. Before initializing, the WBAN data are collected. In the first phase, the user authentication is verified. For this purpose, the user's iris images are aggregated. These iris images are subjected to the Residual Attention Network (RAN). From the RAN, the user is authorized, and then security keys are given to the authorized user. Only after verifying the authentication of the user, the healthcare data is allowed to be stored in the blockchain. In the second phase, data sanitization takes place. The obtained WBAN medical data are sanitized using a data sanitization process with the optimal keys obtained from the Fusion of Golden Eagle and Eurasian Oystercatcher Optimization Algorithm (FGE-EOOA). Here, the data are encrypted by employing the Rivest-Shamir-Adleman (RSA) approach, and then encrypted medical data are stored in the blockchain. This ensures multi-step data security, which allows secure storage of WBAN healthcare data in the blockchain. While retrieving the stored data, the user authentication is verified on the user side, as well as in the same RAN model. This is the third phase of the developed model. When the user is proven to be an authorized one, the stored data in the blockchain corresponding to that particular user is retrieved. Using the data restoration process, which is the fourth phase of the developed model, the actual medical data is retrieved. If the user is unauthorized, then no access is provided to them. This ensures a multi-level of security for storing and retrieving data from the blockchain. The security offered by this model is evaluated and validated by contrasting and comparing it with other conventional data transfer methods.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive Decision-Making for Picking and Replenishment in a DQN-Based Hybrid “Parts-To-Picker” Order Picking System","authors":"Xin Wang,&nbsp;Yaohua Wu","doi":"10.1002/ett.70277","DOIUrl":"https://doi.org/10.1002/ett.70277","url":null,"abstract":"<div>\u0000 \u0000 <p>Some large distribution centers have introduced a hybrid “parts-to-picker” order picking system consisting of a pallet warehouse and a tote warehouse to meet diverse order requirements. This system enables collaborative operations between two warehouses during picking and centralized replenishment processes. Additionally, it innovatively allows surplus goods remaining on pallets after picking to be replenished into the tote warehouse. Therefore, making informed decisions during operations such as picking, centralized replenishment, and picking replenishment will significantly enhance overall warehouse operational efficiency. However, to the best of our knowledge, such research has not yet been conducted. This paper addresses the comprehensive decision-making problem of replenishment and picking in a hybrid “parts-to-picker” order picking system. To solve it, we propose an intelligent decision-making framework based on deep reinforcement learning (DQN). We design a state space that incorporates predictive orders, composite warehouse inventory, and warehouse unit status. Furthermore, we design an action space that includes centralized replenishment, picking replenishment, and picking actions. This approach ultimately achieves three objectives: the allocation of quantities for pallet and tote picking, the allocation of quantities for centralized replenishment, and decisions on whether to replenish after picking. The DQN model also combines a reward function that includes penalty factors with an <span></span><math></math>-greedy decay strategy, effectively improving the goal of order processing efficiency. The experimental results show that, compared with traditional scheduling strategies and intelligent algorithms, the decision-making model trained by the DQN architecture proposed in this paper can respond quickly in the comprehensive decision-making of picking and replenishment in a hybrid “parts-to-picker” order picking system, and can significantly improve system efficiency. The DQN model improves efficiency by approximately 44% and 17% compared to empirical decision-making and meta-heuristic algorithms, respectively. This study provides a solution that combines theoretical innovation and engineering feasibility for the multi-functional collaborative optimization of smart warehouse logistics systems, demonstrating significant practical value.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lightbioptimum: An Intrusion Detection System Based on Bio-Inspired Algorithm for VANET","authors":"Arnaldo Rafael Câmara Araújo,&nbsp;Renata Lopes Rosa,&nbsp;Demóstenes Zegarra Rodríguez,&nbsp;Siti Sarah Maidin,&nbsp;Joseph Bamidele Awotunde,&nbsp;Muhammad Saadi","doi":"10.1002/ett.70254","DOIUrl":"https://doi.org/10.1002/ett.70254","url":null,"abstract":"<div>\u0000 \u0000 <p>In recent years, the development of machine learning-based Intrusion Detection Systems (IDS) has gained significant traction for enhancing data security and identifying threats across diverse network environments. This paper presents a novel lightweight Network Intrusion Detection System (NIDS), named <i>LightBioptimum</i>, specifically designed for Vehicular Ad Hoc Networks (VANETs)—a domain marked by high mobility, dynamic topology, and real-time constraints. The proposed system integrates a bio-inspired optimization technique, Ant Colony Optimization, with a Tree-based Convolutional Neural Network (Tree-CNN) to enable efficient feature selection and accurate threat classification. Experimental evaluations demonstrate that LightBioptimum achieves outstanding results, surpassing existing models in both accuracy and computational efficiency. Notably, it achieves an F1-score of 97.0% in detecting Distributed Denial of Service (DDoS) attacks, outperforming the Deep Belief Network (DBN), which reached 93.0%. Furthermore, LightBioptimum reduces the detection time for brute force attacks by 32.59% compared to DBN. These results confirm the effectiveness of the proposed system in meeting the stringent performance requirements of VANET environments. As Mobile Edge Computing (MEC) applications continue to proliferate in urban areas, LightBioptimum stands out as a promising real-time security solution for VANET and MEC infrastructures.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145223812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Machine Learning-Driven MFP Algorithm for Trajectory Anomaly Detection in Vehicular Ad-Hoc Networks","authors":"Siyu Zhang,&nbsp;Bo Su","doi":"10.1002/ett.70261","DOIUrl":"https://doi.org/10.1002/ett.70261","url":null,"abstract":"<div>\u0000 \u0000 <p>Vehicular ad-hoc networks (VANETs) are integral to the realization of intelligent transportation systems (ITS), enabling seamless communication among vehicles. However, VANETs are highly susceptible to trajectory anomalies (TA) arising from malicious activities, sensor malfunctions, or network attacks, which compromise both network security (NS) and traffic management. The high false positive rates (FPR), computational inefficiency, and poor adaptability to adversarial behaviors may result from the implementation of conventional anomaly detection techniques. Because the pre-defined threshold values or static (FE) Feature Extraction are mostly utilized by those conventional anomaly detection methods. For TA detection in VANET, a novel machine learning-based mobility feature prediction (ML-MFP) algorithm was suggested in this study, and this suggested method is effective to resolve those above-mentioned issues. For the purpose of analyzing vehicular mobility features, including speed, direction, and position, this study also presents the hybrid approach that integrates the supervised learning (SL) with deep learning (DL) techniques with dynamic clustering (DC). In real-time (RT), the prediction of mobility patterns and deviation detection are effectively executed by the application of recurrent neural network (RNN) with long short-term memory (LSTM) units. For high-speed vehicular backgrounds, this method has become effective by this implementation. The issues related to computational overhead, data privacy, and scalability in dense urban networks can be effectively resolved by the integration of lightweight model optimization strategies and privacy-preserving (PP) federated learning (FL). Extensive simulations using real-world vehicular datasets demonstrate that the proposed ML-MFP algorithm achieves high performance, with 98.57% detection accuracy, 97.53% traffic management efficiency, 96.84% network security, 97.17% robustness, and 98.74% anomaly detection. The results validate the effectiveness and practicality of the proposed approach, which enhances VANET security, improves traffic flow, and contributes to the development of intelligent, secure, and efficient transportation systems.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145223811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic Multi-Objective Optimization in Vehicular Fog Computing With NSGA-II+","authors":"Majdi Sukkar,&nbsp;Rajendrasinh Jadeja,&nbsp;Madhu Shukla,&nbsp;Abdullah Albuali,&nbsp;Shakila Basheer","doi":"10.1002/ett.70260","DOIUrl":"https://doi.org/10.1002/ett.70260","url":null,"abstract":"<div>\u0000 \u0000 <p>Vehicular Fog Computing (VFC) presents a promising paradigm to reduce latency and energy usage through utilization of nearby edge resources by vehicles. Yet, efficient and scalable resource management is still a significant challenge particularly due to dynamic network topologies, resource, and high Quality of Service (QoS) requirements. Traditional metaheuristic methods such as GA and PSO are limited in convergence speed and solution quality under such restrictions. This research introduces Enhanced NSGA-II+, a cutting-edge multi-objective evolutionary model enhancing NSGA-II and NSGA-III through dynamic population adaptation, Pareto-front-leveraged selection, and premature convergence prevention. Experimental comparisons in both common and ultra-dense vehicular settings with up to 1000 vehicles and 2000 tasks show that NSGA-II+ outperforms baseline algorithms by far, reducing average delay by 72.55% (compared to NSGA-II) and 71.75% (vs. NSGA-III), and energy cost by 70.96% (compared to NSGA-II) and 70.75% (compared to NSGA-III). This reinforces how NSGA-II+ addresses both dynamic topologies and resource heterogeneity. Its strong exploration-exploitation trade-off and flexibility render it an appealing solution for real-time, energy-efficient deployment in smart transportation systems.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145224417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cloud-Edge-End Collaborative Dependent Computing Schedule Strategy for Immersive Media","authors":"Xiaoxi Wang,&nbsp;Shujie Yang,&nbsp;Hong Tang,&nbsp;Xueying Li,&nbsp;Wei Wang,&nbsp;Hui Xiao,&nbsp;Yuxing Liu,&nbsp;Jia Chen,&nbsp;Enbo Wang,&nbsp;Shaoyun Wu,&nbsp;Mingyu Zhao","doi":"10.1002/ett.70247","DOIUrl":"https://doi.org/10.1002/ett.70247","url":null,"abstract":"<div>\u0000 \u0000 <p>Immersive media applications often create an immersive experience for users through head-mounted displays. However, the computing power and storage capacity of terminal devices are limited, and the local computing architecture cannot meet the high resolution and low latency requirements of panoramic video frames. As a new computing paradigm, cloud, edge and end collaborative computing architecture selectively schedules computing tasks to cloud servers and edge servers with higher computing power, which can effectively improve computing efficiency. However, for dependent computational tasks, the scheduling of each task needs to consider its previous tasks, network state, and computational resources of different servers. Therefore, how to make computational offloading decisions and resource allocation for dependent tasks is a key issue for collaborative computing architectures. This paper investigates and analyzes the immersive media scenarios and the basic computation offloading strategies, and construct a dependent task model graph and optimization problem model. Based on threshold strategy, greedy strategy of heuristic algorithm and deep reinforcement learning model, a scheduling strategy under collaborative computing architecture is designed to maximize the reward related to delay and cost. Finally, the basic performance of the computational task scheduling strategy based on deep reinforcement learning and greedy policy is verified through simulation experiments. The experimental results show that the algorithm reduces the latency by more than 1.8 ms and increases the timely completion rate by more than <span></span><math></math> relative to several basic scheduling schemes, which can effectively improve the service quality and user experience.</p>\u0000 </div>","PeriodicalId":23282,"journal":{"name":"Transactions on Emerging Telecommunications Technologies","volume":"36 10","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145146837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}